Redshift WLM: "final queue may not contain User Groups or Query Groups" -
i want update wlm configuration redshift cluster, unable make changes , save them due following message displayed:
the following problems must corrected before can save workload configuration:
the final queue may not contain user groups or query groups.
now, obvious solution create new queue no user group specified , give remaining amount of memory adds 100%. that's annoying because adding new queue requires cluster reboot, that's not reason i'm asking question.
my main question need new "non-user" queue explained? change because previously, had 4 queues, each assigned user group, , collective memory allocation 87%. didn't add 100%, , supposedly rest dynamically managed redshift.
now, have no problem creating new queue, see explicit explanation of does/what affects before it. didn't see update on official blog, don't see mentioned in docs, or doc updates (http://docs.aws.amazon.com/redshift/latest/mgmt/document-history.html, https://docs.aws.amazon.com/redshift/latest/dg/doc-history.html) blocking me making other desired changes existing queues.
i not sure statement of yours:
previously, had 4 queues, each assigned user group
however, easy guess why redshift requires default queue (i.e. 1 doesn't have user or query groups) last one. let's have 4 queues, each of has different user group, say, ug1, ug2, ug3 , ug4. now, user user group, ug5, queries redshift. queue routed to? default queue helps in such case.
you might counter-argue redshift can use remaining 13% memory. if 100 concurrent queries came ug5. redshift doesn't have configuration of how you'd 13% memory allocated these queries. if manages memory on own, performance of cluster unpredictable.
but yeah, agree redshift documentation lacks clarity on this. closest find this. though, doesn't default queue can't have user or query groups.
the default queue must last queue in wlm configuration. queries not routed other queues run in default queue.
anyway, you'll have live it, guess.
